Enhancing Social Network Trust with Improved EigenTrust Algorithm

In today's online environment, platforms like Facebook and Amazon rely heavily on trust and reputation management systems to ensure their integrity and security. This paper improves the EigenTrust algorithm, a notable model in trust calculation, by incorporating methods from the distributed Bellman Ford algorithm and adaptive weighting to evaluate the trustworthiness of nodes in social networks. Our approach considers both direct and indirect network connections, incorporating feedback credibility. Through extensive experimental analysis, we demonstrate that our modified EigenTrust algorithm (L-level) excels in trust-based P2P systems, significantly outperforming traditional EigenTrust. It effectively reduces unauthentic downloads and maintains high success rates in environments with many malicious collectives, demonstrating robust scalability and reliability as the network expands. This research provides new insights into reputation systems in online environments and suggests directions for future progress in trust management with social networks.